AutoML Leaderboard

Best model	name	model_type	metric_type	metric_value	train_time
	1_Baseline	Baseline	auc	0.5	0.92
	2_DecisionTree	Decision Tree	auc	0.517931	6.86
	3_Linear	Linear	auc	0.549794	3.99
	4_Default_Xgboost	Xgboost	auc	0.556674	51.49
	5_Default_NeuralNetwork	Neural Network	auc	0.46388	2.39
	6_Default_RandomForest	Random Forest	auc	0.566262	12.5
the best	Ensemble	Ensemble	auc	0.577829	0.47

AutoML Performance

AutoML Performance Boxplot

Features Importance

Spearman Correlation of Models

Summary of 5_Default_NeuralNetwork

<< Go back

Neural Network

• n_jobs: -1
• dense_1_size: 32
• dense_2_size: 16
• learning_rate: 0.05
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

1.5 seconds

Metric details

	score	threshold
logloss	0.73664	nan
auc	0.46388	nan
f1	0.498361	0.0152774
accuracy	0.615721	0.505195
precision	0.344828	0.347606
recall	1	0.0152774
mcc	0.0278636	0.347606

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.73664	nan
auc	0.46388	nan
f1	0.137255	0.505195
accuracy	0.615721	0.505195
precision	0.269231	0.505195
recall	0.0921053	0.505195
mcc	-0.0476126	0.505195

Confusion matrix (at threshold=0.505195)

	Predicted as 0	Predicted as 1
Labeled as 0	134	19
Labeled as 1	69	7

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of 1_Baseline

<< Go back

Baseline Classifier (Baseline)

• n_jobs: -1
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

0.2 seconds

Metric details

	score	threshold
logloss	0.635509	nan
auc	0.5	nan
f1	0.498361	0.300437
accuracy	0.331878	0.300437
precision	0.331878	0.300437
recall	1	0.300437
mcc	0	0.300437

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.635509	nan
auc	0.5	nan
f1	0.498361	0.300437
accuracy	0.331878	0.300437
precision	0.331878	0.300437
recall	1	0.300437
mcc	0	0.300437

Confusion matrix (at threshold=0.300437)

	Predicted as 0	Predicted as 1
Labeled as 0	0	153
Labeled as 1	0	76

Learning curves

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of Ensemble

<< Go back

Ensemble structure

Model	Weight
3_Linear	1
4_Default_Xgboost	3
6_Default_RandomForest	2

Metric details

	score	threshold
logloss	0.638378	nan
auc	0.577829	nan
f1	0.510638	0.335527
accuracy	0.689956	0.537046
precision	0.777778	0.537046
recall	1	0.249642
mcc	0.191527	0.537046

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.638378	nan
auc	0.577829	nan
f1	0.164706	0.537046
accuracy	0.689956	0.537046
precision	0.777778	0.537046
recall	0.0921053	0.537046
mcc	0.191527	0.537046

Confusion matrix (at threshold=0.537046)

	Predicted as 0	Predicted as 1
Labeled as 0	151	2
Labeled as 1	69	7

Learning curves

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of 2_DecisionTree

<< Go back

Decision Tree

• n_jobs: -1
• criterion: gini
• max_depth: 3
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

5.9 seconds

Metric details

	score	threshold
logloss	0.745524	nan
auc	0.517931	nan
f1	0.498361	0.113445
accuracy	0.650655	0.517571
precision	0.40625	0.41044
recall	1	0.113445
mcc	0.0636555	0.41044

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.745524	nan
auc	0.517931	nan
f1	0.130435	0.517571
accuracy	0.650655	0.517571
precision	0.375	0.517571
recall	0.0789474	0.517571
mcc	0.0250989	0.517571

Confusion matrix (at threshold=0.517571)

	Predicted as 0	Predicted as 1
Labeled as 0	143	10
Labeled as 1	70	6

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

SHAP Importance

<< Go back

Summary of 6_Default_RandomForest

<< Go back

Random Forest

• n_jobs: -1
• criterion: gini
• max_features: 0.9
• min_samples_split: 30
• max_depth: 4
• eval_metric_name: auc
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

11.4 seconds

Metric details

	score	threshold
logloss	0.64123	nan
auc	0.566262	nan
f1	0.498361	0.0843244
accuracy	0.689956	0.523772
precision	0.608696	0.523772
recall	1	0.0843244
mcc	0.19643	0.523772

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.64123	nan
auc	0.566262	nan
f1	0.282828	0.523772
accuracy	0.689956	0.523772
precision	0.608696	0.523772
recall	0.184211	0.523772
mcc	0.19643	0.523772

Confusion matrix (at threshold=0.523772)

	Predicted as 0	Predicted as 1
Labeled as 0	144	9
Labeled as 1	62	14

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

SHAP Importance

<< Go back

Summary of 4_Default_Xgboost

<< Go back

Extreme Gradient Boosting (Xgboost)

• n_jobs: -1
• objective: binary:logistic
• eta: 0.075
• max_depth: 6
• min_child_weight: 1
• subsample: 1.0
• colsample_bytree: 1.0
• eval_metric: auc
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

50.5 seconds

Metric details

	score	threshold
logloss	0.682156	nan
auc	0.556674	nan
f1	0.498361	0.422421
accuracy	0.68559	0.51789
precision	0.666667	0.51789
recall	1	0.422421
mcc	0.167192	0.51789

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.682156	nan
auc	0.556674	nan
f1	0.181818	0.51789
accuracy	0.68559	0.51789
precision	0.666667	0.51789
recall	0.105263	0.51789
mcc	0.167192	0.51789

Confusion matrix (at threshold=0.51789)

	Predicted as 0	Predicted as 1
Labeled as 0	149	4
Labeled as 1	68	8

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

SHAP Importance

<< Go back

Summary of 3_Linear

<< Go back

Logistic Regression (Linear)

• n_jobs: -1
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

3.0 seconds

Metric details

	score	threshold
logloss	0.686749	nan
auc	0.549794	nan
f1	0.508834	0.0960719
accuracy	0.676856	0.519582
precision	0.538462	0.690156
recall	1	0.0162395
mcc	0.188675	0.519582

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.686749	nan
auc	0.549794	nan
f1	0.362069	0.519582
accuracy	0.676856	0.519582
precision	0.525	0.519582
recall	0.276316	0.519582
mcc	0.188675	0.519582

Confusion matrix (at threshold=0.519582)

	Predicted as 0	Predicted as 1
Labeled as 0	134	19
Labeled as 1	55	21

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

SHAP Importance

<< Go back